Hello, my name is Patrick Langechuan Liu, and I currently work as a detector physicist at the Medical Imaging division of PerkinElmer Inc. I received my PhD with distinction in Physics at the University of Michigan in Ann Arbor.
Trained as a physicist, I have developed strong analytical skills and a particular interest in imaging science. I have constantly found that the most enjoyable part of my job involves gaining useful insights through mining large sets of data.
• Department of Physics, Rackham Graduate School
• Department of Radiation Oncology - Physics Division
• Department of Physics, School of Physics
• Center for Theoretical Biology
• Performed design and test various amorphous silicon and CMOS detectors for X-ray imaging
• Optimized detector test procedures and cut testing time by 60% through analysis of 5 years of test data (60 TB+ image and text files stored on in-house server) in Python
• Reduced time for defect analysis from 5 minutes to 3 seconds (100x speedup) by developing new algorithms
• Developed a GUI to analyze test images and identify detectors of passing grade using logistic regression
• Increased X-ray detector spatial resolution by 80%+ by adopting a barrier rib structure in the scintillator layer
• Selected as one of 30 fellows from 1800 applicants to participate in an intensive data science fellowship program
• Performed sentiment analysis through construction of bag-of-words and bigram models based on more than 1 million Yelp reviews using Scikit-learn library in Python
• Predicted business star-rating through linear regression of the features extracted from 30,000+ Yelp records
• Identified the most common restaurant violations in different cuisine types based on analysis of more than 500,000 restaurant inspection records in New York City using SQL
• Constructed and analyzed a social network of celebrities in NYC using data scraped from 1200+ webpages
• Revealed relationships among Wikipedia pages by analyzing the network of links on all English Wikipedia pages (10 GB+ of HTML files stored on Amazon Web Service) using MapReduce algorithms in mrjob framework
• Pioneered design optimization of megavoltage X-ray detectors for radiotherapy portal imaging and MV CBCT
• Designed and conducted numerical modeling through parallel Monte Carlo simulation on 800+ CPU cores
• Invented a hybrid modeling framework to reduce simulation time from 100 million down to only 30 CPU hours
• Designed and implemented CT reconstruction algorithm and imaging performance analysis in MATLAB
• Created and implemented model-based iterative reconstruction algorithms for limited field-of-view CT imaging
• Communicated and collaborated closely with researchers and consultants across research institutes
• Led a team of three students and coordinated teamwork on an affordable motorized commuter bike
• Conducted over 50 interviews, pivoted prototype design and business model based on customer discovery
• Established partnership with occupational education program at local community college
I was selected as one of the three U of M participants for a on-site program offered by Boston Consulting Group for Advanced Degree holders
• Analyzed and created a detalied solution for a mock consulting case
• Presented the solution in front of executives and fellow participants
I was selected as one of the two U of M participants for a on-site program offered by McKinsey for Advanced Degree holders
• Obtained the problem-solving methodology in management consulting industry
• Team-worked through practices of mini-cases, group discussion and presentation
Daolan Translation Group is the largest and most famous non-profit documentary translation group in China (http://daolan.net/)
• Lead 7 projects to translate feature-length documentaries from Japanese to Chinese
• Bridged information gap due to language barrier and television program censorship
• Collaborated online with university professors and professional translators from all over the world
• Work highlighted by media and adopted by more than 20 universities as educational materials
• Delivered independently prepared physics lectures to over 150 university students
• Tutored Mechanics and Electromagnetism physics labs, graded lab reports and research project papers
• Received excellent reviews from students during course feedback
• Initiated a customized T-shirt design contest to promote Peking University Comics Club
• Reduced cost by 50% through creation of online ordering tools and negotiation for competitive quotes
• Profited $300 from 1000+ orders, which was donated to Relief Fund for the 2008 Sichuan Earthquake
Langechuan Liu, Larry Antonuk, Youcef El-Mohri, Hao Jiang, Qihua Zhao, Medical Physics (Accepted, 2015)
Langechuan Liu, Larry Antonuk, Youcef El-Mohri, Qihua Zhao, Hao Jiang, Medical Physics, 41, 061916 (2014)
Youcef El-Mohri, Larry Antonuk, Richard Choroszucha, Qihua Zhao, Hao Jiang, Langechuan Liu, Physics in Medicine and Biology, 59 (2014) 797-818
Langechuan Liu, Larry Antonuk, Qihua Zhao, Youcef El-Mohri, Hao Jiang Physics in Medicine and Biology, 57 (2012) 5343-58
Youcef El-Mohri, Larry Antonuk, Qihua Zhao, Richard Choroszucha, Hao Jiang, Langechuan Liu Physics in Medicine and Biology, 56 (2011) 1509-1527
Larry Antonuk, Langechuan Liu, Albert Liang, Youcef El-Mohri, Qihua Zhao, Martin Koniczek, Hao Jiang, SPIE Medical Imaging 2015, 9412-14
Larry Antonuk, Langechuan Liu, Youcef El-Mohri, Hao Jiang, Qihua Zhao, RSNA 2014, SSJ21-06
Langechuan Liu, Larry Antonuk, Youcef El-Mohri, Qihua Zhao, Hao Jiang, AAPM 2014, TH-A-18C-1
Langechuan Liu, Larry Antonuk, Hao Jiang, Youcef El-Mohri, Qihua Zhao, RSNA 2013, SST15-06
Larry Antonuk, Youcef El-Mohri, Qihua Zhao, Langechuan Liu, Hao Jiang, RSNA 2012, LL-PHS-TH2B
Larry Antonuk, Langechuan Liu, Qihua Zhao, Youcef El-Mohri, Hao Jiang, Robert Street AAPM 2011, Radiography/Projection Imaging Section, SU-C-220-6
Qihua Zhao, Langechuan Liu, Youcef El-Mohri, Larry Antonuk, Hao Jiang, Martin Koniczek AAPM 2010, Imaging General Section, SU-GG-I-136
